rehydra

Overview

rehydra scans LLM prompts for personally identifiable information (PII) before they reach the model. It acts as a pre-processing step to redact or block sensitive data, helping developers avoid accidental leakage. The tool is written in TypeScript and integrates into existing workflows.

Best for

Best for
Developers building LLM applications that handle user data and need to prevent PII leakage

Use cases

• Sanitizing user inputs before sending to an LLM API
• Redacting PII from prompts in customer support chatbots
• Ensuring compliance with data privacy regulations in LLM workflows

Notes

rehydra scans LLM prompts for personally identifiable information (PII) before they reach the model. It acts as a pre-processing step to redact or block sensitive data, helping developers avoid accidental leakage. The tool is written in TypeScript and integrates into existing workflows.

66 stars on GitHub. Last updated 2026-06-01. Licensed MIT.

Use cases

• Sanitizing user inputs before sending to an LLM API
• Redacting PII from prompts in customer support chatbots
• Ensuring compliance with data privacy regulations in LLM workflows

Pros

• Simple integration into existing TypeScript projects
• Reduces risk of exposing sensitive data
• Lightweight and focused on a single task

Cons

• Limited to TypeScript or JavaScript environments
• May not catch all edge cases of PII (e.g., context-dependent)
• Requires manual configuration of detection rules

Indexed from awesome-generative-ai and enriched against its public facts.